US 20110205340A1

(19) United States (12) Patent Application Publication (10) Pub. No.: US 2011/0205340 A1

Garcia et al. (43) Pub. Date: Aug. 25, 2011

(54) 3D TIME-OF-FLIGHT CAMERA SYSTEM (30) Foreign Application Priority Data AND POSITION/ORIENTATION CALIBRATION METHOD THEREFOR Aug. 12, 2008 (EP) ................................ .. 081622227

(75) Inventors: Frederic Garcia, Palafrugell Publication Classi?cation

(GGir°‘:}3i.(ES)S; nitric (FR) (51) Int. Cl. ran 1 ier, a1n - ax ; Bruno Mirbach, KonZ (DE); H04N 13/02 (200601) Roberto 01-56110, Luxembourg (52) US. Cl. ................................. .. 348/46; 348/E13.074

(LU); Thomas Solignac, Luttange (FR) (57) ABSTRACT

(73) Assignee IEE INTERNATIONAL A camera system comprises a 3D TOF camera for acquiring ' ELECTRONICS & a camera-perspectlve range 1mage~of a scene ‘and an image

ENGINEERING SIA- processor'for processlng the range lmage. The image proces ECHTERNACH (LU) ’ sor conta1ns a pos1t1on and onentatlon cal1brat1on routine

implemented therein in hardware and/or software, Which (21) AP p1~ N O‘: 13/058,962 position and orientation calibration routine, When executed

by the image processor, detects one or more planes Within a (22) PCT Filed, API._ 27, 2009 range image acquired by the 3D TOF camera, selects a refer

ence plane among the at least one or more planes detected and (86) PCT NO; pCT/Ep2009/055054 computes position and orientation parameters of the 3D TOF

camera With respect to the reference plane, such as, e. g., § 371 (0X1), elevation above the reference plane and/or camera roll angle (2), (4) Date; May 6, 2011 and/or camera pitch angle.

Rotation (R) and translation (if)

14..’

in

Patent Application Publication Aug. 25, 2011 Sheet 1 0f 3 US 2011/0205340 A1

Fig. 1

Rotation (R) and translation (I)

{W}

W Fig. 2

Patent Application Publication Aug. 25, 2011 Sheet 2 0f 3 US 2011/0205340 A1

Fig. 3

Patent Application Publication Aug. 25, 2011 Sheet 3 0f 3 US 2011/0205340 A1

Fig. 4

Mount camera

I “H Input approximate camera orientation

w

Acquire range image

r W Compute Cartesian representatiun of scene

in camera reference system

, we Piane detection

J] ‘-\+§ Plane selectien

L ‘WW: Compute camera position

L w? Compute coordinate transformation matrix

\ wry

Cempute vectors I'm

US 2011/0205340 A1

3D TIME-OF-FLIGHT CAMERA SYSTEM AND POSITION/ORIENTATION

CALIBRATION METHOD THEREFOR

TECHNICAL FIELD

[0001] The present invention generally relates to the tech nical ?eld of 3D computer vision, in particular to a position/ orientation calibration method for a 3D time-of-?ight camera system.

BACKGROUND

[0002] As used herein, a 3D time-of-?ight camera (herein after “3D TOF camera” or simply “camera”) designates a camera that acquires images (herein called “range images”) containing distance information pertaining to a scene by using the time-of-?ight measurement principle. Such camera generally comprises a light source emitting modulated or pulsed light into the scene to be imaged in 3D and a pixel array on Which the scene is imaged by an optical system. The camera then correlates the light detected in the pixels With the light emitted and determines, for each pixel, the phase differ ence betWeen emitted and received light, or the difference betWeen emission and reception times. These quantities are proportional to the radial distance betWeen the camera and the part of the scene that is imaged onto the pixel concerned. Some 3D TOF cameras are knoWn that acquire range images in real time. [0003] The ?elds of application of 3D TOF cameras are many and diverse. It has been proposed to employ such 3D TOF cameras in the automotive ?eld, e. g. for passenger detec tion and classi?cation in a vehicle, for monitoring the sur roundings of a car. 3D TOF cameras have also been suggested for controlling access to restricted areas (eg in airports, banks or public buildings), in particular for detecting, track ing and counting of persons and objects. An important advan tage of 3D TOF camera systems is that the depth (or range) information is readily available and does not require cost intensive computation as, for instance, in stereoscopic sys tems.

[0004] To the knowledge of the inventors, position and orientation calibration of 3D TOF cameras currently requires the person installing the camera to go through relatively com plicated procedures, Which might hinder the success of such cameras on the market place. Beraldin et al. present in their paper “Practical Range Camera Calibration” Proc. Soc. Photo-Opt. Instum. Eng. a calibration procedure adapted to a range camera intended for space applications. They propose a tWo-step methodology to perform the calibration, the ?rst step being a speci?c calibration for the close-range volume (from 0.5 m to 1.5 m). In this ?rst step, they use an array of targets positioned at knoWn locations in the ?eld of vieW of the range camera using a precise linear stage. In the second step, they propose to use a smaller number of targets posi tioned at distances greater than 1.5 m With the help of an accurate electronic distance-measuring device. As is depicted in this paper, reference points in the scene are used in order to perform the calibration process. The paper “High Resolution Segmentation With a Time-of-Flight 3D-Camera using the Example of a Lecture” Scene by Santrac et al. (Freie Univer sitat Berlin, 2006) deals With the combination of range images from a 3D TOF camera and images from a 2D camera. To map the images to the tWo cameras, a calibration step is performed prior to using the camera combination for segmen

Aug. 25, 2011

tation. For this purpose, a calibration pattern With at least 5>

US 2011/0205340 A1

angle, pitch angle and elevation of the camera With respect to the reference plane. For the purposes of the present, We Will use the convention that the camera points to the negative camera-Z-axis, and that the camera-x axis and the camera-y axis are oriented perpendicular one another and the camera

Z-axis to the camera. Except for these restrictions, camera-x and camera-y-axes can otherWise be freely selected. In the context of the present, “roll angle” designates an angle mea suring rotation about the camera-x-axis and “pitch angle” designates an angle measuring rotation about the camera-y axis. To the inventors’ best knowledge, this is the ?rst cali bration method that alloWs determining orientation and posi tion parameters of a 3D TOF camera in non-lab conditions by taking pro?t of the raW data provided and Without using prede?ned reference points in the scene (i.e. points that have to be marked in the actual scene in such a Way that they can be identi?ed in the range image). [0012] It is to be noted that for the present method, the internal camera calibration should have been carried out beforehand, e. g. in the camera factory. As used herein, “inter nal calibration” designates calibration of camera parameters, Which are independent of the location Where the camera is installed, e. g. optical system parameters such as focal length, principal point location and lens distortion. A method of internal camera calibration is disclosed, eg in EP 1 662 440 A1. In contrast to internal calibration, the present method addresses “extemal calibration”, i.e. calibration of param eters, Which do depend on the location Where the camera is installed.

[0013] Whatever is the intended ?eld of application of a 3D TOF camera system, the range images provided are prefer ably evaluated automatically. For this purpose, the range images have to be processed by appropriate means, such as eg object or pattern recognition softWare. One of the bene?ts of the present invention is that the same softWare can be used on range images provided by 3D TOF cameras mounted in different locations, since method proposed alloWs compen sating for position and/or orientation differences betWeen the cameras.

[0014] A further aspect of the invention concerns a camera system con?gured and arranged for carrying out the position and orientation calibration method. Such system comprises a 3D TOF camera for acquiring a camera-perspective range image of a scene and an image processor for processing the range image. The image processor contains a position and orientation calibration routine implemented therein in hard Ware and/or softWare, Which position and orientation calibra tion routine, When executed by the image processor, detects one or more planes Within a range image acquired by the 3D TOF camera, selects a reference plane among the at least one or more planes detected and computes position and orienta tion parameters of the 3D TOF camera With respect to the reference plane, such as, e.g., elevation above the reference plane and/or camera roll angle and/or camera pitch angle. [0015] The detection of one or more planes Within the range image may be carried out using RANSAC-based plane detec tion. RANSAC (RANdom SAmple Consensus) is an algo rithm for ?tting of models, Which is particularly robust against outliers. A short description of the RANSAC algo rithm is given in the paper “Random Sample Consensus (RANSAC)” by H. CantZler (University of Edinburgh), a more detailed one is given in M. A. Fischler and R. C. Bolles: “Random sample consensus: A paradigm for model ?tting

Aug. 25, 2011

With application to image analysis and automated cartogra phy”, Communications of the ACM, 24(6):381-395, 1981. [0016] Whether a RANSAC algorithm is used or not, the detection of one or more planes may advantageously com prise compensating for range errors induced by light spread ing in the 3D TOF camera and/or discarding image pixels containing range information deemed unreliable (e.g. pixels having too little brighness or outliers). A preferred method for compensating the effects of light spreading is described in European patent application 07 024 505.5. [0017] Preferably, the selection of a reference plane out of the planes detected comprises identifying a ?oor plane and ?xing the ?oor plane as the reference plane. It is expected that in most applications the ?oor Will lie in the ?eld of vieW of the camera, since surveillance cameras are commonly mounted in an elevate position above and directed doWnWard to the area to be monitored. Most preferably, the selection of the reference plane is effected autonomously by the calibration routine based upon and folloWing input of user-de?ned limits of camera roll angle and/ or camera pitch angle With respect to the ?oor plane. It may happen that the scene imaged by the camera contains more than one planar surface (?oor and Walls for instance). In this case, the person Who installs the camera may input rough orientation indications (like: “roll angle between —450 and +450 ”, “pitch angle betWeen —90° and 0° ”, etc.) in order to alloW the camera to determine the ?oor plane autonomously. [0018] The selection of the reference plane may also com prise the presentation of the one or more detected planes using a user interface and ?xing of the reference plane based upon user interaction.

[0019] If the method requires some interaction With the user, the camera system preferably comprising a user inter face for presenting output data to and/or receiving input data from a user. Such user interface may, for instance, comprise a display screen, a keyboard or a keypad, a touchscreen, etc. As an alternative, the camera system could comprise a hardWare interface alloWing a user to temporarily connect a user inter face (eg a laptop) for setting up the system. Preferably, a user interface comprises a display screen enabling the person installing the camera to visualiZe the scene in the camera perspective. If the person also has to select a plane, the one or more planes detected could be indicated on the display screen (eg in different colours). [0020] Preferably, the calibration routine comprises com puting coordinate transformation parameters of the coordi nate transformation betWeen the camera coordinate system and the World coordinate system. Such coordinate transfor mation transforms, When applied to a camera-perspective range image of the scene, such range image into a Cartesian representation of the scene, in Which coordinates are de?ned With respect to the reference plane. The camera system pref erably comprises a memory, in Which the coordinate trans formation parameters may be stored. [0021] As Will be appreciated, the camera system as pre sented herein is suitable, in particular for pedestrian detection at crossWalks, since the scene imaged in such application typically comprises an extended ?oor plane that may serve as the reference plane.

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Further details and advantages of the present inven tion Will be apparent from the folloWing detailed description of not limiting embodiments With reference to the attached draWings, Wherein:

US 2011/0205340 A1

[0023] FIG. 1 is a schematic diagram illustrating the rela tionship between World and the camera coordinate systems; [0024] FIG. 2 is a schematic illustration of a 3D TOF cam era system; [0025] FIG. 3 is a perspective vieW of a crossWalk surveil lance system including a 3D TOF camera; [0026] FIG. 4 is a How diagram of a preferred method of installing a 3D TOF camera.

DETAILED DESCRIPTION

[0027] FIG. 2 shoWs a schematic illustration of a 3D TOF camera system 10, comprising a 3D TOF camera 12 and an image processor 14 (shoWn as an element separate from the camera 12, but Which may also be incorporated With it in the same housing). The camera comprises an optical imaging system, illustrated by lens 16, and an illumination unit, illus trated by LED array 18, for illuminating the scene to be imaged With modulated or pulsed light. The camera system also comprises a hardWare interface, eg a USB port, a ?reWire port, etc. (not shoWn in the draWing) alloWing the connection of a user interface (in this case a laptop computer

20) [0028] Range images acquired With a 3D TOF camera cor respond to matrices of distance values d, Which indicate the distances from the camera to the imaged surface elements. Given the internal camera parameters of the camera (e.g. focal length, principal point of the camera, and distortion parameters), the 3D coordinates C[X,Y,Z]T of the visible sur face elements Within the scene can be determined With refer ence to the camera coordinate system. Coordinates expressed With reference to the camera coordinate system are preceded With a superscript “c” for distinction With coordinates

Aug. 25, 2011

uses the convention that the rotations are performed clock Wise, ?rst around the Z-axis, then around the y-axis already once rotated, and ?nally around the x-axis that is already tWice rotated during the previous stages:

Where the elementary rotation matrices [RZ], [Ry] and [Rx] are given by:

cosa —sinz1 O (3)

[R1] : sina cosa O] O O 1

005,3 0 sin? (4) [Ry] = O l O

—sin,B 0 005,3

1 O O (5)

[RX] : O cosy —siny

O siny cosy

[0031] By substituting the rotation matrices (3,4,5) in (2), the resulting rotation matrix that de?nes camera orientation is:

cosacos? cosasin?sirry- cosasin?cosy+ (6) sinacosy sinasiny

expressed in the World coordinate system (i.e. With respect to cosacosy cosasiny the reference plane), Which are preceded by a superscript “W”. —sin,B cos ,Bsiny cos?cosy In a possible embodiment this is achieved by computing the product betWeen the acquired distance value at each pixel and a unit vector C[ex,ey,eZ]T de?ned according to the internal [0032] Hence, from (1):

WX cosacos? cosasin?siny — sinacosy cosasin?cosy + sinasiny IX EX (7)

WY sinacos? sinasin?sirry + cosacosy sinasin?cosy — cosasiny [y CY

WZ : —sin,B cos?siny cos?cosy [Z CZ

1 0 0 0 1 1

camera parameters for each pixel. This computation is [0033] To simplify the notation, We denote the matrix ele explained in EP 1 662 440 Al, incorporated herein by refer ence in its entirety.

[0029] As far as the relationship betWeen the World coor dinate system and the camera coordinate system {C} is composed of a rotation matrix (3x3) and a translation vector (as illustrated in FIG. 1), the folloWing equation can be de?ned:

WMIZRIFM (1)

Where M represents a coordinate vector augmented by l as the last element M:[X,Y,Z,l]T to alloW the homography repre sentation betWeen the camera and the World coordinate sys tems.

[0030] This rotation matrix R is presented using Euler angles 0t, [3 and y that de?ne a sequence of three elementary rotations around Z , y and x-axis respectively. One normally

ment in the ith roW and the jth column of the rotation matrix R by alj. From (7):

WX an 6112 6113 IX EX (8)

WY an an 6123 [y CY

WZ 6131 6132 6133 I1 CZ

1 0 0 0 l 1

[0034] As long as no external calibration of the camera has been performed, only the coordinates of the 3D scene points With reference to the camera coordinate system, ie C[X,Y,Z], are known. In order to determine the transformation matrix [R t] betWeen the camera and the World coordinate system, it Will be necessary to ?x some constraints on the data.

US 2011/0205340 A1

[0035] In the following, we will assume that the scene imaged by the camera contains the ?oor and that it has been selected in the calibration routine as the reference plane. [0036] The fact of having selected a plane in the scene allows us to ?x the Z coordinate of the points contained in that plane, with reference to the world coordinate system, to a speci?c value. For convenience, this coordinate will be ?xed to 0. Thus, considering WZIO, it follows from (8):

[0037] The last equation of the system of equations in (9) is the one of interest. This equation presents 4 unknown vari ables [a31 a32 a33 tZ] that implies that at least 4 points [cXk, CYk, cZk], k:1, . . . n belonging to the ?oor are necessary to solve the following system of equations:

where n; 14. The more points coming from the ?oor, the more accurate will be the value of each unknown variable. [0038] One way to ef?ciently solve the system in (10) is by writing (10) in matrix form

0 cXl cYl 0Z1 1 6131 (11) : _ : : : : £132

[ 0 ] EX” CY” CZ” 1

[0039] and applying singular value decomposition (SVD), which is described, for instance, in “Matrix Computations”, 3rd edition, 1996, Johns Hopkins University Press, Baltimore, by Golub et al. [0040] From the SVD computation, the singular vector that corresponds to its smallest singular value gives the least square approximation to its solution, but up to a scale factor called 7». Therefore, from the SVD computation, instead of [a31 a32 a33 tZ] values, what results is [v1 v2 v3 v4], with:

mat-v2

Mat-v3

lz:7vv4 (12)

[0041] In order to ?x the correct value for the unknown variables [a3 1 a32 a33] it is necessary to determine the 7» value. Since the unknown variables correspond to the last rotation matrix row and from one of the orthogonal rotation matrix properties:

RT-RII (13)

where I represents the unity matrix, the following constraint can be de?ned:

a312+a322+a332:1 (14)

Aug. 25, 2011

[0042] The scale factor 7» can be determined arranging the equations in (12) in the previous constraint (14) as follows:

[0043] There a two possible solutions, differing by the sign. For the following, the solution of 7» will be considered, for which the matrix coe?icient a33 (see equation (12)) is posi tive. This solution corresponds to the situation where the normal vector of the ?oor corresponds to the Z-axis of the world coordinate system, such that the origin of the camera reference system has a positive Z-component in the world coordinate system. [0044] Alternatively to the SVD computation, a direct Least Square computation involving the constraint of equa tion (14) can be computed, eg as is presented in “Direct Least Square Fitting of Ellipses”, Tern Analysis and Machine Intelligence, Vol. 2 1 , No. 5, May 1999, by FitZgibbon et al. By computing the scatter matrix of the transformation matrix between the camera coordinate system and the world coordi nate system in (11) a 4x4 matrix is obtained. This matrix can be combined with the constraining matrix of (14), yielding a generalised eigensystem that has to be solved. The lower computational cost of the direct Least Squares method in comparison to the SVD makes it suitable for implementation on real microprocessors.

[0045] Once the value of 7» is ?xed, the unknown variables can be automatically estimated by substituting 7» in (12). What is thus directly determined is the translation tZ in Z-direction as well as the 3rd row of the rotation matrix (6):

a3] : —sin,B (16)

r132 : cos?siny

[0046] Equation (16) can be solved for obtaining the rota tion angles [3 and y as follows:

,3 : arcsin(—a31) (17)

_ E (l 3) y _ arctan[ £133 )

[0047] Notice that there are two possible values for [3 and y leading to the same solution. Limiting the range of the angle y to a half-circle, eg to the interval [—J'l§/2,J'IZ/2] the solution becomes unambiguous. This constraint would means that the camera is assumed to be rotated around the x-axis less than 90° from the initial orientation in which the camera coordi nate system coincides with the world coordinate system.

[0048] From these computations, angles [3 and y, which correspond to pitch and roll rotations, respectively, have been estimated. The remaining angle to complete the rotation matrix is the yaw angle 0t, which can be ?xed to an arbitrary value, eg 0° for convenience. The translation vectort can be ?xed to [0,0,tZ]T which which means that the origin of the world coordinate system lies vertically underneath the origin

US 2011/0205340 A1

of the camera coordinate system. Thus, the transformation matrix presented in (1) may be fully determined using the ?oor as the reference plane. [0049] The method as presented so far in this detailed description corresponds to that part of the external camera calibration, Which folloWs the selection of the reference plane. In the following, We Will address the detection of planes in the scenes and the selection of a reference plane in more detail. When objects (eg fumiture or people) and/or more than one plane (roof or Walls) are present in the scene, robust ?tting routines have to be used to detect the plane or planes Within the scene. In order to distinguish betWeen points in the range image belonging to a plane and points belonging to an object, a RANSAC based plane detection is preferably used. The RANSAC algorithm is an iterative algo rithm for robust ?tting of models. It is robust in the sense of good tolerance to outliers in the experimental data. Its struc ture is simple but poWerful, based on the folloWing steps: ?rst, the algorithm randomly selects three points of the range image and computes the number of inliers belonging to the plane (With some error tolerance) spanned by the three points. The algorithm then goes on by selecting three neW points and making the same operations until no more data can be ?tted on a plane or until a prede?ned number of iterations is reached. Those three points that de?ne the best plane(s) (With the highest number of inliers) Will be returned by the algo rithm. The planes de?ned by the different point triplets are not necessarily the best ?t to the planes actually present in the scene. Accordingly, a better estimation of the plane param eters may be computed using eg a Least Squares method. As Will be appreciated, points belonging to objects or persons in the scene Will normally be discarded since they do not belong to a plane.

[0050] It cannot be assumed a priori that only one plane is present in the scene. A strategy has therefore to be adopted for selecting a reference plane out of the planes detected. If only one plane is detected, in most surveillance applications one may assume With a reasonable degree of certainty that this plane is a ?oor plane and the calibration routine could ?xing that plane as the reference plane. If more than one plane is detected some input should be provided to the calibration routine by the person installing the camera. [0051] If the camera is looking to the ?oor, the ?oor can be discriminated from other planes in the scene (eg the ceiling or Walls) as folloWs: the camera is installed above the ?oor, the estimated Z-coordinate tZ of the camera in the World coor dinate system is positive (Where it is assumed that the World coordinate system is so oriented that points above the ?oor level have a positive Z-coordinate). Any ceiling above the camera can easily be discarded from the set of candidates of the ?oor plane by imposing the condition tZ>0 to the detected planes for being further considered. By requiring the pitch angle [3 around the y-axis to lie in the interval [—J'|Z/ 4,J1:/ 4], the

condition |a31|:|sin [3|

US 2011/0205340 A1

image of the scene to be monitored is then acquired in step 43. Provided that the camera orientation has not been changed in the meanwhile, a range image acquired beforehand (When the camera Was mounted) could also be used for the calibration and step 43 be omitted.

[0057] The calibration routine then calculates the coordi nates of the points of the range image in the camera coordi nate system (in step 44), eg using the method proposed in EP 1 662 440 Al, Whereby the optical parameters of the camera (known beforehand and stored in a memory of the camera) are taken into account. In EP 1 662 440 Al, the optical camera parameters are stored in form of a set of vectors, each vector belonging to a speci?c pixel of the range image. The coordi nate calculation method of this step comprises, for each pixel, multiplying the range value determined for the pixel With that stored vector that corresponds to the speci?c pixel:

cx; Get-X (19) C)’; = d; - Get-y

CZ; ceiz

Where d,- is the distance measured in the i-th pixel, [cex? ceyi, CeZi]T is the vector stored for the i-th pixel, and the left-hand side is the coordinate vector of the image point in the camera reference system. The result of this step is a Cartesian repre sentation of the scene in the camera coordinate system.

[0058] The next steps are plane detection (at reference numeral 45), Which as described above, is preferably carried out using a RANSAC algorithm and selection of the reference plane among the planes detected (step 46). As mentioned, ?oor plane identi?cation may be done autonomously (i.e. Without user interaction at this stage), using eg the previ ously de?ned approximate orientation of the camera, or based upon user interaction (e.g. requesting the user to indicate the ?oor plane). At this stage of the procedure, the calibration routine has su?icient information to compute camera position (in terms of vertical distance from the reference plane) and orientation (roll and pitch angles With respect to the reference plane), Which is thus carried out in step 47. As explained above, camera position and orientation translates into a coor dinate transformation from the camera coordinate system into the World reference system. Accordingly, in step 48, the cali bration routine computes this coordinate transformation (eg in form of a matrix) and combines it With the above coordinate calculation method, Whereby a set of vectors [Wex? Weyi, WeZi]T is obtained in step 49, by means of Which the distance values measured for each pixel may be directly converted into a point expressed by its coordinates in the World reference system:

62-‘ (20)

Where R is given by equation (2). It should be noted that in a practical embodiment of the routine, the steps 48 and 49 could be implemented as a single step. From the range image (i.e. the set of pixels With distance values d,), the coordinates of the points in the World reference system can be calculated by:

Aug. 25, 2011

[X (21)

[0059] Once the calibration routine has been successfully executed, the data necessary for transforming the range images acquired by the camera into a Cartesian representa tion of the scene in the World coordinate system are the vectors [wex? Weyi, WeZi]T and [tx, ty, tZ]T. Nevertheless, the camera preferably keeps the vectors [cex? eyi, CeZi]T in memory in case the calibration has to be repeated (e. g. after servicing the camera). [0060] As those skilled Will appreciate, the present inven tion is suited for most applications of 3D TOF cameras and greatly facilitates installation of such cameras in a real-World environment.

1-15. (canceled) 16. Position and orientation calibration method for a cam

era system including a 3D time-of-?ight camera, said method comprising

acquiring a camera-perspective range image of a scene using said 3D time-of-?ight camera,

detecting one or more planes Within said range image and selecting a reference plane among said one or more

planes detected, computing position and orientation parameters of said 3D

time-of-?ight camera With respect to said reference plane.

17. The method as claimed in claim 16, Wherein said posi tion and orientation parameters of said 3D time-of ?ight camera include at least one of height above said reference plane, camera roll angle and camera pitch angle.

18. The method as claimed in claim 16, Wherein said detec tion of one or more planes Within said range image comprises RANSAC-based plane detection.

19. The method as claimed in claim 16, Wherein said select ing of a reference plane comprises identifying a ?oor plane and ?xing said ?oor plane as said reference plane.

20. The method as claimed in claim 19, Wherein said selec tion of said reference plane is effected by said camera based upon and folloWing input of user-de?ned limits of at least one of camera roll angle and camera pitch angle With respect to said ?oor plane.

21. The method as claimed in claim 16, Wherein said selec tion of said reference plane comprises presenting said one or more detected planes using a user interface and ?xing said reference plane based upon user interaction.

22. The method as claimed in claim 16, comprising computing coordinate transformation parameters of a

coordinate transformation, said coordinate transforma tion being such that it transforms, When applied to a camera-perspective range image of said scene, such range image into a Cartesian representation of said scene, in Which coordinates are de?ned With respect to said reference plane,

and storing said coordinate transformation parameters Within a memory of said camera.

23. The method as claimed in claim 16, Wherein said detec tion of one or more planes comprises at least one of compen sating for range errors induced by light spreading in said 3D

US 2011/0205340 A1

time-of-?ight camera and discarding image pixels containing range information deemed unreliable.

24. A camera system comprising a 3D time-of-?ight camera for acquiring a camera-perspec

tive range image of a scene, and an image processor for processing said range image, Wherein said image processor comprises a position and

orientation calibration routine implemented in at least one of hardWare and software, Wherein said position and orientation calibration routine, When executed by said image processor, detects one or more planes Within a range image acquired by said 3D time-of-?ight camera, selects a reference plane among said one or more planes detected and computes position and orientation param eters of said 3D time-of-?ight camera With respect to said reference plane.

25. The camera system according to claim 24, Wherein said position and orientation parameters comprise at least one of height above said reference plane, camera roll angle and camera pitch angle.

26. The camera system as claimed in any one of claims 24, comprising a user interface for presenting output data to or receiving input data from a user.

27. The camera system as claimed in claim 24, Wherein said position and orientation calibration routine, When executed by said image processor, selects said reference plane by identifying a ?oor plane and ?xing said ?oor plane as said reference plane.

28. The camera system as claimed in claim 26, comprising a user interface for presenting output data to and receiving input data from a user, Wherein said calibration routine, When executed by said image processor, selects said reference plane based upon and folloWing input, via said user interface, of user-de?ned limits of at least one of camera roll angle and camera pitch angle With respect to said ?oor plane.

Aug. 25, 2011

29. The camera system as claimed in claim 25, Wherein said selection of said reference plane comprises presenting said one or more detected planes using a user interface and ?xing said reference plane based upon user interaction.

30. The camera system as claimed in claim 24, comprising a memory, and Wherein said calibration routine, When executed by said image processor, computes coordinate trans formation parameters of a coordinate transformation from a camera reference system into a World reference system, in Which coordinates are de?ned With respect to said reference plane, and stores said coordinate transformation parameters Within said memory.

31. Pedestrian detection system comprising a camera sys tem as claimed in claim 24.

32. A camera system comprising a 3D time-of-?ight camera for acquiring a camera-perspec

tive range image of a scene, and an image processor for processing said range image, Wherein said image processor is con?gured to detect one or

more planes Within a range image acquired by said 3D time-of-?ight camera, to select a reference plane among said one or more planes detected and to compute posi tion and orientation parameters of said 3D time-of-?ight camera With respect to said reference plane, Wherein said camera system comprises a memory,

and Wherein said processor is further con?gured to com pute coordinate transformation parameters of a coordi nate transformation from a camera reference system into a World reference system, in Which coordinates are de?ned With respect to said reference plane, to store said coordinate transformation parameters Within said memory.